Data analyses summary
Environment preparation
Importation data and functions
BMIQ_all: TCGA methylation data with IDHm, IDHWT and CD34+
BMIQ_CD34: Methylome from GSE103008, CD34+ Methylation
BMIQ_Wang_Feng: Methylome from Koichi’s study
BMIQ_Whiele: Methylome from Whiele’s study
Wang_Feng_RNAseq: Gene expression from Koichi’s study
Preparation of annotations data
Loading of gene promoter annotation from Blueprint project
Loading CpGs annotation from Illumina manifest
Loading chromatin conformation annotation
Creation of Granges object with coordinates
Overlapping of the different sources of annotations
gene_list <- c("CEBPA", "CPT1A", "CPT2", "SLC25A20", "AKT2", "PPARGC1A", "PTEN")
Blueprint <- read.csv("../BLUEPRINT_fragments_good.tsv", sep = "\t") %>%
dplyr::select(., "gene_names":"type", "gene_type") %>%
separate_rows(., gene_names, sep = " ") %>%
separate_rows(., gene_type, sep = " ") %>%
unique(.)
pchic <- prepare_pchic()
PCHiC_bed <- unique(rbind(pchic[, c(1:3, 5)], pchic[, c(6:8, 10)]))
PCHiC_GRange <- GRanges(
seqnames = PCHiC_bed$chr,
IRanges(start = PCHiC_bed$start, end = PCHiC_bed$end),
Gene_Pchic = PCHiC_bed$Name,
start_fragment = PCHiC_bed$start,
end_fragment = PCHiC_bed$end
)
PCHiC_GRange$ID <- paste(PCHiC_bed$chr, PCHiC_bed$start, sep = "_")
colnames(pchic) <- c("chr_bait", "start_bait", "end_bait", "ID_bait", "Name_bait", "chr_oe", "start_oe", "end_oe", "ID_oe", "Name_oe")
pchic$IDbait <- paste(pchic$chr_bait, pchic$start_bait, sep = "_")
pchic$IDoe <- paste(pchic$chr_oe, pchic$start_oe, sep = "_")
Blueprint_Granges <- GRanges(
seqnames = Blueprint$chr,
ranges = IRanges(start = Blueprint$start, end = Blueprint$end),
Blueprint_gene_names = Blueprint$gene_names,
type = Blueprint$type,
gene_type = Blueprint$gene_type
)
## ILLUMINA 450K
anno_450 <- read.csv("~/Illumina_Manifest/HumanMethylation450_15017482_v1-2.csv", skip = 7) %>%
dplyr::select(., "Name", "CHR", "MAPINFO", "UCSC_RefGene_Name", "UCSC_RefGene_Group", "Relation_to_UCSC_CpG_Island") %>%
dplyr::filter(., CHR != "") %>%
separate_rows(., UCSC_RefGene_Name, UCSC_RefGene_Group, sep = ";")
CpGs_Granges_450 <- GRanges(
seqnames = anno_450$CHR,
ranges = IRanges(anno_450$MAPINFO, anno_450$MAPINFO +1),
CpG = anno_450$Name,
Illumina_Gene_name = anno_450$UCSC_RefGene_Name,
position = anno_450$UCSC_RefGene_Group,
Island = anno_450$Relation_to_UCSC_CpG_Island
)
overlaps_CpGs_Blueprint_450 <- findOverlaps(Blueprint_Granges, CpGs_Granges_450)
match_hit_CpGs_Blueprint_450 <- data.frame(mcols(Blueprint_Granges[queryHits(overlaps_CpGs_Blueprint_450),]),
data.frame(mcols(CpGs_Granges_450[subjectHits(overlaps_CpGs_Blueprint_450),])))
match_hit_CpGs_Blueprint_promoter_450 <- dplyr::filter(match_hit_CpGs_Blueprint_450, type == "P")
anno_promoter_450 <- dplyr::filter(anno_450, UCSC_RefGene_Group == "TSS1500" | UCSC_RefGene_Group == "TSS200")
overlaps_CpGs_Pchic_450 <- findOverlaps(CpGs_Granges_450, PCHiC_GRange)
matchit_CpGs_Pchic_450 <- data.frame(mcols(CpGs_Granges_450[queryHits(overlaps_CpGs_Pchic_450),]),
data.frame(mcols(PCHiC_GRange[subjectHits(overlaps_CpGs_Pchic_450),])))
gene_universe_450K <- unique(match_hit_CpGs_Blueprint_promoter_450$Blueprint_gene_names)
#########################
## ILLUMINA EPIC
anno_EPIC <- read.csv("~/Illumina_Manifest/infinium-methylationepic-v-1-0-b5-manifest-file.csv", skip = 7) %>%
dplyr::select(., "Name", "CHR", "MAPINFO", "UCSC_RefGene_Name", "UCSC_RefGene_Group", "Relation_to_UCSC_CpG_Island") %>%
dplyr::filter(., CHR != "") %>%
separate_rows(., UCSC_RefGene_Name, UCSC_RefGene_Group, sep = ";")
CpGs_Granges_EPIC <- GRanges(
seqnames = anno_EPIC$CHR,
ranges = IRanges(anno_EPIC$MAPINFO, anno_EPIC$MAPINFO +1),
CpG = anno_EPIC$Name,
Illumina_Gene_name = anno_EPIC$UCSC_RefGene_Name,
position = anno_EPIC$UCSC_RefGene_Group,
Island = anno_EPIC$Relation_to_UCSC_CpG_Island
)
overlaps_CpGs_Blueprint_EPIC <- findOverlaps(Blueprint_Granges, CpGs_Granges_EPIC)
match_hit_CpGs_Blueprint_EPIC <- data.frame(mcols(Blueprint_Granges[queryHits(overlaps_CpGs_Blueprint_EPIC),]),
data.frame(mcols(CpGs_Granges_EPIC[subjectHits(overlaps_CpGs_Blueprint_EPIC),])))
match_hit_CpGs_Blueprint_promoter_EPIC <- dplyr::filter(match_hit_CpGs_Blueprint_EPIC, type == "P")
anno_promoter_EPIC <- dplyr::filter(anno_EPIC, UCSC_RefGene_Group == "TSS1500" | UCSC_RefGene_Group == "TSS200")
overlaps_CpGs_Pchic_EPIC <- findOverlaps(CpGs_Granges_EPIC, PCHiC_GRange)
matchit_CpGs_Pchic_EPIC <- data.frame(mcols(CpGs_Granges_EPIC[queryHits(overlaps_CpGs_Pchic_EPIC),]),
data.frame(mcols(PCHiC_GRange[subjectHits(overlaps_CpGs_Pchic_EPIC),])))
gene_universe_EPIC <- unique(match_hit_CpGs_Blueprint_promoter_EPIC$Blueprint_gene_names)
##########################
overlaps_Blueprint_Pchic <- findOverlaps(Blueprint_Granges, PCHiC_GRange)
matchit_Blueprint_Pchic <- data.frame(mcols(Blueprint_Granges[queryHits(overlaps_Blueprint_Pchic),]),
data.frame(mcols(PCHiC_GRange[subjectHits(overlaps_Blueprint_Pchic),]))) %>%
dplyr::filter(., type == "P")
signature <- read.csv("../DATA/BPmetCan.txt", sep = "\t")
rownames(signature) <- signature$CpGs
signature <- signature[,-1]TCGA
Preparation of the data
Phenotype_BMIQ_CD34_IDHm_WT <- read.csv("../../TCGA_Connections/Phenotype_met.csv") %>%
dplyr::filter(., WT1 == "WT1WT", DNMT3A == "DNMT3AWT", TET2 == "TET2WT", FLT3 == "FLT3AWT")
Phenotype_BMIQ_CD34_IDHm_WT$Phenotype <- ifelse(Phenotype_BMIQ_CD34_IDHm_WT$IDH == "IDHm", "IDHm", "WT")
Phenotype_BMIQ_CD34_IDHm_WT <- dplyr::select(Phenotype_BMIQ_CD34_IDHm_WT, c(X, Phenotype)) %>%
rbind(data.frame(X = colnames(BMIQ_CD34), Phenotype = rep("CD34", 6)))
Phenotype_BMIQ_CD34_IDHm_WT$X <- str_replace(Phenotype_BMIQ_CD34_IDHm_WT$X, "-", ".")
Phenotype_BMIQ_CD34_IDHm_WT$X <- str_replace(Phenotype_BMIQ_CD34_IDHm_WT$X, "-", ".")
BMIQ_CD34_IDHm_WT <- BMIQ_all %>%
dplyr::select(., Phenotype_BMIQ_CD34_IDHm_WT$X)Global data
Deconvolution using methylation data and BPmetCan signature
res <- epidish(as.matrix(BMIQ_CD34_IDHm_WT), as.matrix(signature), method = "RPC")
Fres <- as.data.frame(res$estF)
Fres <- cbind(Sample = rownames(Fres), Fres)
Fres[, -1] <- round(Fres[, -1], 3)
deconvolution_violin_TCGA_DATA <- data.frame(Cell_Type = rep(colnames(Fres[,c(2:12)]), each = length(rownames(Fres))),
Values = Fres[,c(2:12)] %>% c(.) %>% unlist(.) %>% as.vector(.))
deconvolution_violin_TCGA_DATA$Values <- round(deconvolution_violin_TCGA_DATA$Values, 1)
ggplot(deconvolution_violin_TCGA_DATA, aes(x=Cell_Type, y=Values, fill=Cell_Type), las = 2) +
geom_violin() +
ggtitle("TCGA all sample")+
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))
DNA methylation state
Most variable CpGs
BMIQ_all_high_variance <- Focus_high_variance_CpGs(BMIQ_CD34_IDHm_WT, percentage = 1)
Focus_global_methylation(match_hit_CpGs_Blueprint_450, anno_450, Phenotype_BMIQ_CD34_IDHm_WT, BMIQ_all_high_variance, c("IDHm", "WT"), fill_colour = c("#00FF00", "#FF0000", "#0000FF", "#00FF00"))
Promoter of genes of interest
Focus_Gene_global_methylation("CEBPA", match_hit_CpGs_Blueprint_promoter_450, anno_promoter_450, Phenotype_BMIQ_CD34_IDHm_WT, BMIQ_CD34_IDHm_WT, c("IDHm", "WT"), c("#00FF00", "#FF0000", "#0000FF"))
Focus_Gene_global_methylation("CPT1A", match_hit_CpGs_Blueprint_promoter_450, anno_promoter_450, Phenotype_BMIQ_CD34_IDHm_WT, BMIQ_CD34_IDHm_WT, c("IDHm", "WT"), c("#00FF00", "#FF0000", "#0000FF"))
Focus_Gene_global_methylation("CPT2", match_hit_CpGs_Blueprint_promoter_450, anno_promoter_450, Phenotype_BMIQ_CD34_IDHm_WT, BMIQ_CD34_IDHm_WT, c("IDHm", "WT"), c("#00FF00", "#FF0000", "#0000FF"))
Focus_Gene_global_methylation("SLC25A20", match_hit_CpGs_Blueprint_promoter_450, anno_promoter_450, Phenotype_BMIQ_CD34_IDHm_WT, BMIQ_CD34_IDHm_WT, c("IDHm", "WT"), c("#00FF00", "#FF0000", "#0000FF"))
Focus_Gene_global_methylation("PPARGC1A", match_hit_CpGs_Blueprint_promoter_450, anno_promoter_450, Phenotype_BMIQ_CD34_IDHm_WT, BMIQ_CD34_IDHm_WT, c("IDHm", "WT"), c("#00FF00", "#FF0000", "#0000FF"))
Focus_Gene_global_methylation("AKT2", match_hit_CpGs_Blueprint_promoter_450, anno_promoter_450, Phenotype_BMIQ_CD34_IDHm_WT, BMIQ_CD34_IDHm_WT, c("IDHm", "WT"), c("#00FF00", "#FF0000", "#0000FF"))
Focus_Gene_global_methylation("PTEN", match_hit_CpGs_Blueprint_promoter_450, anno_promoter_450, Phenotype_BMIQ_CD34_IDHm_WT, BMIQ_CD34_IDHm_WT, c("IDHm", "WT"), c("#00FF00", "#FF0000", "#0000FF"))
Neighbor of promoter of genes of interest
Focus_gene_promoter_neighborhood_methylation("CEBPA", match_hit_CpGs_Blueprint_promoter_450, anno_promoter_450, Phenotype_BMIQ_CD34_IDHm_WT, BMIQ_CD34_IDHm_WT, c("IDHm", "WT"), c("#00FF00", "#FF0000", "#0000FF"), matchit_CpGs_Pchic_450, pchic, match_hit_CpGs_Blueprint_450)
Focus_gene_promoter_neighborhood_methylation("CPT1A", match_hit_CpGs_Blueprint_promoter_450, anno_promoter_450, Phenotype_BMIQ_CD34_IDHm_WT, BMIQ_CD34_IDHm_WT, c("IDHm", "WT"), c("#00FF00", "#FF0000", "#0000FF"), matchit_CpGs_Pchic_450, pchic, match_hit_CpGs_Blueprint_450)
Focus_gene_promoter_neighborhood_methylation("CPT2", match_hit_CpGs_Blueprint_promoter_450, anno_promoter_450, Phenotype_BMIQ_CD34_IDHm_WT, BMIQ_CD34_IDHm_WT, c("IDHm", "WT"), c("#00FF00", "#FF0000", "#0000FF"), matchit_CpGs_Pchic_450, pchic, match_hit_CpGs_Blueprint_450)
Focus_gene_promoter_neighborhood_methylation("SLC25A20", match_hit_CpGs_Blueprint_promoter_450, anno_promoter_450, Phenotype_BMIQ_CD34_IDHm_WT, BMIQ_CD34_IDHm_WT, c("IDHm", "WT"), c("#00FF00", "#FF0000", "#0000FF"), matchit_CpGs_Pchic_450, pchic, match_hit_CpGs_Blueprint_450)## [1] "No Cpgs found in neighbor of promoter of SLC25A20"## [1] FALSEFocus_gene_promoter_neighborhood_methylation("PPARGC1A", match_hit_CpGs_Blueprint_promoter_450, anno_promoter_450, Phenotype_BMIQ_CD34_IDHm_WT, BMIQ_CD34_IDHm_WT, c("IDHm", "WT"), c("#00FF00", "#FF0000", "#0000FF"), matchit_CpGs_Pchic_450, pchic, match_hit_CpGs_Blueprint_450)## [1] "No Cpgs found in neighbor of promoter of PPARGC1A"## [1] FALSEFocus_gene_promoter_neighborhood_methylation("AKT2", match_hit_CpGs_Blueprint_promoter_450, anno_promoter_450, Phenotype_BMIQ_CD34_IDHm_WT, BMIQ_CD34_IDHm_WT, c("IDHm", "WT"), c("#00FF00", "#FF0000", "#0000FF"), matchit_CpGs_Pchic_450, pchic, match_hit_CpGs_Blueprint_450)
Focus_gene_promoter_neighborhood_methylation("PTEN", match_hit_CpGs_Blueprint_promoter_450, anno_promoter_450, Phenotype_BMIQ_CD34_IDHm_WT, BMIQ_CD34_IDHm_WT, c("IDHm", "WT"), c("#00FF00", "#FF0000", "#0000FF"), matchit_CpGs_Pchic_450, pchic, match_hit_CpGs_Blueprint_450)
Gene Ontology
IDHm vs WT
BMIQ_TCGA_analysis <- Differential_analysis(Phenotype_BMIQ_CD34_IDHm_WT$Phenotype, BMIQ_CD34_IDHm_WT)## 1 done## 2 done## 3 doneResponse_diff_hypermet_in_IDHm <- BMIQ_TCGA_analysis[["IDHm-WT"]] %>%
dplyr::filter(., P.Value < 0.01) %>%
dplyr::filter(., logFC > 0.3)
Response_diff_hypermet_in_WT <- BMIQ_TCGA_analysis[["IDHm-WT"]] %>%
dplyr::filter(., P.Value < 0.01) %>%
dplyr::filter(., logFC < 0.3)
Genes_hypermetylated_in_IDHm <- Look_at_gene_with_CpGs_in_promoter(Response_diff_hypermet_in_IDHm$ID, match_hit_CpGs_Blueprint_promoter_450)
Genes_hypermetylated_in_IDHm_enhancer <- Look_at_genes_connected_to_CpGs(Response_diff_hypermet_in_IDHm$ID,
matchit_CpGs_Pchic_450,
pchic,
matchit_Blueprint_Pchic,
match_hit_CpGs_Blueprint_promoter_450)
Genes_hypermetylated_in_WT <- Look_at_gene_with_CpGs_in_promoter(Response_diff_hypermet_in_WT$ID, match_hit_CpGs_Blueprint_promoter_450)
Genes_hypermetylated_in_WT_enhancer <- Look_at_genes_connected_to_CpGs(Response_diff_hypermet_in_WT$ID,
matchit_CpGs_Pchic_450,
pchic,
matchit_Blueprint_Pchic,
match_hit_CpGs_Blueprint_promoter_450)
Genes_hypermetylated_in_IDHm_ego <- enrichGO(
gene = Genes_hypermetylated_in_IDHm,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_450K
)
Genes_hypermetylated_in_IDHm_enhancer_ego <- enrichGO(
gene = Genes_hypermetylated_in_IDHm_enhancer,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_450K
)
Genes_hypermetylated_in_WT_ego <- enrichGO(
gene = Genes_hypermetylated_in_WT,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_450K
)
Genes_hypermetylated_in_WT_enhancer_ego <- enrichGO(
gene = Genes_hypermetylated_in_WT_enhancer,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_450K
)
Genes_hypermetylated_in_IDHm_promoter_and_enhancer_ego <- enrichGO(
gene = unique(c(Genes_hypermetylated_in_IDHm, Genes_hypermetylated_in_IDHm_enhancer)),
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_450K
)
Genes_hypermetylated_in_WT_promoter_and_enhancer_ego <- enrichGO(
gene = unique(c(Genes_hypermetylated_in_WT, Genes_hypermetylated_in_WT_enhancer)),
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_450K
)
dotplot(Genes_hypermetylated_in_IDHm_ego, showCategory = 30, title = "Hypermet IDHm", font.size = 6)
dotplot(Genes_hypermetylated_in_IDHm_enhancer_ego, showCategory = 30, title = "Hypermet Neighborhood IDHm")
dotplot(Genes_hypermetylated_in_IDHm_promoter_and_enhancer_ego, showCategory = 30, title = "Prom & enhancer IDHm")
dotplot(Genes_hypermetylated_in_WT_ego, showCategory = 30, title = "Hypermet WT")
dotplot(Genes_hypermetylated_in_WT_enhancer_ego, showCategory = 30, title = "Hypermet Neighborhood WT")
dotplot(Genes_hypermetylated_in_WT_promoter_and_enhancer_ego, showCategory = 30, title = "Prom & enhancer WT")
Volcano plots
IDHm vs IDH WT
volcanoplot_methylation(BMIQ_TCGA_analysis[["IDHm-WT"]], match_hit_CpGs_Blueprint_450, "IDHm vs WT")
Whiele’s DATA
Global data
Principal Component Analysis
DATA_for_PCA <- t(as.data.frame(BMIQ_Whiele))
res.pca <- PCA(DATA_for_PCA, ncp = 3, graph = TRUE)
fviz_eig(res.pca, addlabels = TRUE)
Clustering
BMIQ_Whiele_high_variance <- Focus_high_variance_CpGs(BMIQ_Whiele, percentage = 1)
Phenotype_Whiele$Phenotype <- ifelse(Phenotype_Whiele$Phenotype == "WT+2HG", "WTHG", Phenotype_Whiele$Phenotype)
ann_color <- list(
Phenotype = c(IDH2m = "red", WT = "blue", Control = "grey", WTHG = "green"))
Make_heatmap(BMIQ_Whiele, Phenotype_Whiele, "pearson", "All values", ann_color)
Make_heatmap(BMIQ_Whiele_high_variance, Phenotype_Whiele, "pearson", "High variance", ann_color)
Deconvolution
res <- epidish(as.matrix(BMIQ_Whiele), as.matrix(signature), method = "RPC")
Fres <- as.data.frame(res$estF)
Fres <- cbind(Sample = rownames(Fres), Fres)
Fres[, -1] <- round(Fres[, -1], 3)
deconvolution_violin_Whiele_DATA <- data.frame(Cell_Type = rep(colnames(Fres[,c(2:12)]), each = length(rownames(Fres))),
Values = Fres[,c(2:12)] %>% c(.) %>% unlist(.) %>% as.vector(.))
deconvolution_violin_Whiele_DATA$Values <- round(deconvolution_violin_Whiele_DATA$Values, 1)
ggplot(deconvolution_violin_Whiele_DATA, aes(x=Cell_Type, y=Values, fill=Cell_Type), las = 2) +
geom_violin() +
ggtitle("All the sample")+
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))
DNA methylation state
Most variable CpGs
Focus_global_methylation(match_hit_CpGs_Blueprint_EPIC, anno_EPIC, Phenotype_Whiele, BMIQ_Whiele, c("IDH2m", "WT"), c("#555555", "#FF0000", "#0000FF", "#00FF00"))
Promoter of genes of interest
Focus_Gene_global_methylation("CEBPA", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Whiele, BMIQ_Whiele, c("IDH2m", "WT"), c("#555555", "#FF0000", "#0000FF", "#00FF00"))
Focus_Gene_global_methylation("CPT1A", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Whiele, BMIQ_Whiele, c("IDH2m", "WT"), c("#555555", "#FF0000", "#0000FF", "#00FF00"))
Focus_Gene_global_methylation("CPT2", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Whiele, BMIQ_Whiele, c("IDH2m", "WT"), c("#555555", "#FF0000", "#0000FF", "#00FF00"))
Focus_Gene_global_methylation("SLC25A20", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Whiele, BMIQ_Whiele, c("IDH2m", "WT"), c("#555555", "#FF0000", "#0000FF", "#00FF00"))
Focus_Gene_global_methylation("PPARGC1A", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Whiele, BMIQ_Whiele, c("IDH2m", "WT"), c("#555555", "#FF0000", "#0000FF", "#00FF00"))
Focus_Gene_global_methylation("AKT2", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Whiele, BMIQ_Whiele, c("IDH2m", "WT"), c("#555555", "#FF0000", "#0000FF", "#00FF00"))
Focus_Gene_global_methylation("PTEN", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Whiele, BMIQ_Whiele, c("IDH2m", "WT"), c("#555555", "#FF0000", "#0000FF", "#00FF00"))
Neighbor of promoter of genes of interest
Focus_gene_promoter_neighborhood_methylation("CEBPA", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Whiele, BMIQ_Whiele, c("IDH2m", "WT"), c("#555555", "#FF0000", "#0000FF", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)
Focus_gene_promoter_neighborhood_methylation("CPT1A", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Whiele, BMIQ_Whiele, c("IDH2m", "WT"), c("#555555", "#FF0000", "#0000FF", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)
Focus_gene_promoter_neighborhood_methylation("CPT2", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Whiele, BMIQ_Whiele, c("IDH2m", "WT"), c("#555555", "#FF0000", "#0000FF", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)
Focus_gene_promoter_neighborhood_methylation("SLC25A20", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Whiele, BMIQ_Whiele, c("IDH2m", "WT"), c("#555555", "#FF0000", "#0000FF", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)## [1] "No Cpgs found in neighbor of promoter of SLC25A20"## [1] FALSEFocus_gene_promoter_neighborhood_methylation("PPARGC1A", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Whiele, BMIQ_Whiele, c("IDH2m", "WT"), c("#555555", "#FF0000", "#0000FF", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)## [1] "No Cpgs found in neighbor of promoter of PPARGC1A"## [1] FALSEFocus_gene_promoter_neighborhood_methylation("AKT2", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Whiele, BMIQ_Whiele, c("IDH2m", "WT"), c("#555555", "#FF0000", "#0000FF", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)
Focus_gene_promoter_neighborhood_methylation("PTEN", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Whiele, BMIQ_Whiele, c("IDH2m", "WT"), c("#555555", "#FF0000", "#0000FF", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)
Gene Ontology
IDH2 vs WT
BMIQ_Whiele_analysis <- Differential_analysis(Phenotype_Whiele$Phenotype, BMIQ_Whiele)## 1 done## 2 done## 3 done## 4 done## 5 done## 6 doneResponse_diff_hypermet_in_IDH2m <- BMIQ_Whiele_analysis[["IDH2m-WT"]] %>%
dplyr::filter(., P.Value < 0.01) %>%
dplyr::filter(., logFC > 0.3)
Response_diff_hypermet_in_WT_whiele <- BMIQ_Whiele_analysis[["IDH2m-WT"]] %>%
dplyr::filter(., P.Value < 0.01) %>%
dplyr::filter(., logFC < 0.3)
Genes_hypermetylated_in_IDH2m <- Look_at_gene_with_CpGs_in_promoter(Response_diff_hypermet_in_IDH2m$ID, match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_IDH2m_enhancer <- Look_at_genes_connected_to_CpGs(Response_diff_hypermet_in_IDH2m$ID,
matchit_CpGs_Pchic_EPIC,
pchic,
matchit_Blueprint_Pchic,
match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_WT_whiele <- Look_at_gene_with_CpGs_in_promoter(Response_diff_hypermet_in_WT_whiele$ID, match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_WT_enhancer_whiele <- Look_at_genes_connected_to_CpGs(Response_diff_hypermet_in_WT_whiele$ID,
matchit_CpGs_Pchic_EPIC,
pchic,
matchit_Blueprint_Pchic,
match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_IDH2m_ego <- enrichGO(
gene = Genes_hypermetylated_in_IDH2m,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_IDH2m_enhancer_ego <- enrichGO(
gene = Genes_hypermetylated_in_IDH2m_enhancer,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_WT_whiele_ego <- enrichGO(
gene = Genes_hypermetylated_in_WT_whiele,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_WT_enhancer_ego_whiele <- enrichGO(
gene = Genes_hypermetylated_in_WT_enhancer_whiele,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_IDH2m_promoter_and_enhancer_ego <- enrichGO(
gene = unique(c(Genes_hypermetylated_in_IDH2m, Genes_hypermetylated_in_IDH2m_enhancer)),
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_WT_promoter_and_enhancer_ego_whiele <- enrichGO(
gene = unique(c(Genes_hypermetylated_in_WT_whiele, Genes_hypermetylated_in_WT_enhancer_whiele)),
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
dotplot(Genes_hypermetylated_in_IDH2m_ego, showCategory = 30, title = "Hypermet IDHm")
dotplot(Genes_hypermetylated_in_IDH2m_enhancer_ego, showCategory = 30, title = "Hypermet Neighborhood IDHm", font.size = 9)
dotplot(Genes_hypermetylated_in_IDH2m_promoter_and_enhancer_ego, showCategory = 30, title = "Prom & enhancer IDHm")
dotplot(Genes_hypermetylated_in_WT_whiele_ego, showCategory = 30, title = "Hypermet WT")
dotplot(Genes_hypermetylated_in_WT_enhancer_ego_whiele, showCategory = 30, title = "Hypermet Neighborhood WT", font.size = 9)
dotplot(Genes_hypermetylated_in_WT_promoter_and_enhancer_ego_whiele, showCategory = 30, title = "Prom & enhancer WT")
Volcano plots
IDH2 vs WT
volcanoplot_methylation(BMIQ_Whiele_analysis[["IDH2m-WT"]], match_hit_CpGs_Blueprint_EPIC, "IDH2m vs WT")
Koichi’s data Analyses
Global data
Principal Component Analysis
DATA_for_PCA <- t(as.data.frame(BMIQ_Wang_Feng))
res.pca <- PCA(DATA_for_PCA, ncp = 3, graph = TRUE)
fviz_eig(res.pca, addlabels = TRUE)
Clustering
All samples
BMIQ_Wang_Feng_high_variance <- Focus_high_variance_CpGs(BMIQ_Wang_Feng, percentage = 1)
ann_color <- list(
Phenotype = c(Baseline = "blue", Non_Responder = "red", Responder = "green", Control = "grey"))
Make_heatmap(BMIQ_Wang_Feng, Phenotype_Wang_Feng, "pearson", "All values", ann_color)
Make_heatmap(BMIQ_Wang_Feng_high_variance, Phenotype_Wang_Feng, "pearson", "High variance", ann_color)
Control/Baseline and Responder/Non-responder
Phenotype_Wang_Feng_Control_Baseline <- Phenotype_Wang_Feng %>%
dplyr::filter(., Phenotype == "Baseline" | Phenotype == "Control")
BMIQ_Control_Baseline <- dplyr::select(BMIQ_Wang_Feng, Phenotype_Wang_Feng_Control_Baseline$Sample_number)
BMIQ_Control_Baseline_high_variance <- Focus_high_variance_CpGs(BMIQ_Control_Baseline, 1)
Phenotype_Wang_Feng_Responder_Non_Responder <- Phenotype_Wang_Feng %>%
dplyr::filter(., Phenotype != "Baseline" & Phenotype != "Control")
BMIQ_Responder_Non_Responder <- dplyr::select(BMIQ_Wang_Feng, Phenotype_Wang_Feng_Responder_Non_Responder$Sample_number)
BMIQ_Responder_Non_Responder_high_variance <- Focus_high_variance_CpGs(BMIQ_Responder_Non_Responder, 1)
Make_heatmap(BMIQ_Control_Baseline, Phenotype_Wang_Feng_Control_Baseline, "pearson", "Control vs Baseline", ann_color)
Make_heatmap(BMIQ_Control_Baseline_high_variance, Phenotype_Wang_Feng_Control_Baseline, "pearson", "Control vs Baseline High variance", ann_color)
Make_heatmap(BMIQ_Responder_Non_Responder, Phenotype_Wang_Feng_Responder_Non_Responder, "pearson", "Responder vs Non-Responder", ann_color)
Make_heatmap(BMIQ_Responder_Non_Responder_high_variance, Phenotype_Wang_Feng_Responder_Non_Responder, "pearson", "Responder vs Non-Responder High variance", ann_color)
Baseline
Phenotype_Wang_Feng_Baseline <- Phenotype_Wang_Feng %>%
dplyr::filter(., Phenotype == "Baseline")
Phenotype_cluster <- read.csv("../DATA/Wang_Feng_DATA/Clustering_Baseline_phenotype.csv")
Phenotype_Wang_Feng_Baseline_clustered <- merge(Phenotype_Wang_Feng_Baseline, Phenotype_cluster, by.x = "Sample", by.y = "UPN", all.x = TRUE)
Phenotype_Wang_Feng_Baseline_clustered$Phenotype <- Phenotype_Wang_Feng_Baseline_clustered$Cluster
BMIQ_Baseline <- dplyr::select(BMIQ_Wang_Feng, Phenotype_Wang_Feng_Baseline_clustered$Sample_number)
BMIQ_Baseline_high_variance <- Focus_high_variance_CpGs(BMIQ_Baseline, 1)
#BMIQ_Baseline_high_variance_promoter <- Focus_high_variance_CpGs(BMIQ_Baseline_promoter, 1)
ann_color <- list(
Phenotype = c(cluster1 = "blue", cluster2 = "red"))
Make_heatmap(BMIQ_Baseline, Phenotype_Wang_Feng_Baseline_clustered, "pearson", "Baseline", ann_color)
Make_heatmap(BMIQ_Baseline_high_variance, Phenotype_Wang_Feng_Baseline_clustered, "pearson", "Baseline High variance", ann_color)
#Make_heatmap(BMIQ_Baseline_promoter, Phenotype_Wang_Feng_Baseline_clustered, "pearson", "Baseline promoter", ann_color)
#Make_heatmap(BMIQ_Baseline_high_variance_promoter, Phenotype_Wang_Feng_Baseline_clustered, "pearson", "Baseline High variance promoter", ann_color)Deconvolution
signature <- read.csv("../DATA/BPmetCan.txt", sep = "\t")
rownames(signature) <- signature$CpGs
signature <- signature[,-1]
res <- epidish(as.matrix(BMIQ_Wang_Feng), as.matrix(signature), method = "RPC")
Fres <- as.data.frame(res$estF)
Fres <- cbind(Sample = rownames(Fres), Fres)
Fres[, -1] <- round(Fres[, -1], 3)
deconvolution_violin_Koichi_Data <- data.frame(Cell_Type = rep(colnames(Fres[,c(2:12)]), each = length(rownames(Fres))),
Values = Fres[,c(2:12)] %>% c(.) %>% unlist(.) %>% as.vector(.))
deconvolution_violin_Koichi_Data$Values <- round(deconvolution_violin_Koichi_Data$Values, 1)
Baseline_sample <- Phenotype_Wang_Feng %>%
.[.$Phenotype == "Baseline", "Sample_number"]
res_Baseline <- epidish(as.matrix(BMIQ_Wang_Feng %>%
dplyr::select(., Baseline_sample)), as.matrix(signature), method = "RPC")## Note: Using an external vector in selections is ambiguous.
## i Use `all_of(Baseline_sample)` instead of `Baseline_sample` to silence this message.
## i See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
## This message is displayed once per session.Fres_Baseline <- as.data.frame(res_Baseline$estF)
Fres_Baseline <- cbind(Sample = rownames(Fres_Baseline), Fres_Baseline)
Fres_Baseline[, -1] <- round(Fres_Baseline[, -1], 3)
deconvolution_violin_Baseline <- data.frame(Cell_Type = rep(colnames(Fres_Baseline[,c(2:12)]), each = length(rownames(Fres_Baseline))),
Values = Fres_Baseline[,c(2:12)] %>% c(.) %>% unlist(.) %>% as.vector(.))
deconvolution_violin_Baseline$Values <- round(deconvolution_violin_Baseline$Values, 1)
Responder_sample <- Phenotype_Wang_Feng %>%
.[.$Phenotype == "Responder", "Sample_number"]
res_Responder <- epidish(as.matrix(BMIQ_Wang_Feng %>%
dplyr::select(., Responder_sample)), as.matrix(signature), method = "RPC")## Note: Using an external vector in selections is ambiguous.
## i Use `all_of(Responder_sample)` instead of `Responder_sample` to silence this message.
## i See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
## This message is displayed once per session.Fres_Responder <- as.data.frame(res_Responder$estF)
Fres_Responder <- cbind(Sample = rownames(Fres_Responder), Fres_Responder)
Fres_Responder[, -1] <- round(Fres_Responder[, -1], 3)
deconvolution_violin_Responder <- data.frame(Cell_Type = rep(colnames(Fres_Responder[,c(2:12)]), each = length(rownames(Fres_Responder))),
Values = Fres_Responder[,c(2:12)] %>% c(.) %>% unlist(.) %>% as.vector(.))
deconvolution_violin_Responder$Values <- round(deconvolution_violin_Responder$Values, 1)
Non_Responder_sample <- Phenotype_Wang_Feng %>%
.[.$Phenotype == "Non_Responder", "Sample_number"]
res_Non_Responder <- epidish(as.matrix(BMIQ_Wang_Feng %>%
dplyr::select(., Non_Responder_sample)), as.matrix(signature), method = "RPC")## Note: Using an external vector in selections is ambiguous.
## i Use `all_of(Non_Responder_sample)` instead of `Non_Responder_sample` to silence this message.
## i See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
## This message is displayed once per session.Fres_Non_Responder <- as.data.frame(res_Non_Responder$estF)
Fres_Non_Responder <- cbind(Sample = rownames(Fres_Non_Responder), Fres_Non_Responder)
Fres_Non_Responder[, -1] <- round(Fres_Non_Responder[, -1], 3)
deconvolution_violin_Non_Responder <- data.frame(Cell_Type = rep(colnames(Fres_Non_Responder[,c(2:12)]), each = length(rownames(Fres_Non_Responder))),
Values = Fres_Non_Responder[,c(2:12)] %>% c(.) %>% unlist(.) %>% as.vector(.))
deconvolution_violin_Non_Responder$Values <- round(deconvolution_violin_Non_Responder$Values, 1)
Control_sample <- Phenotype_Wang_Feng %>%
.[.$Phenotype == "Control", "Sample_number"]
res_Control <- epidish(as.matrix(BMIQ_Wang_Feng %>%
dplyr::select(., Control_sample)), as.matrix(signature), method = "RPC")## Note: Using an external vector in selections is ambiguous.
## i Use `all_of(Control_sample)` instead of `Control_sample` to silence this message.
## i See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
## This message is displayed once per session.Fres_Control <- as.data.frame(res_Control$estF)
Fres_Control <- cbind(Sample = rownames(Fres_Control), Fres_Control)
Fres_Control[, -1] <- round(Fres_Control[, -1], 3)
deconvolution_violin_Control <- data.frame(Cell_Type = rep(colnames(Fres_Control[,c(2:12)]), each = length(rownames(Fres_Control))),
Values = Fres_Control[,c(2:12)] %>% c(.) %>% unlist(.) %>% as.vector(.))
deconvolution_violin_Control$Values <- round(deconvolution_violin_Control$Values, 1)
ggplot(deconvolution_violin_Koichi_Data, aes(x=Cell_Type, y=Values, fill=Cell_Type), las = 2) +
geom_violin() +
ggtitle("All the sample")+
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))
ggplot(deconvolution_violin_Baseline, aes(x=Cell_Type, y=Values, fill=Cell_Type), las = 2) +
geom_violin() +
ggtitle("Baseline")+
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))
ggplot(deconvolution_violin_Responder, aes(x=Cell_Type, y=Values, fill=Cell_Type), las = 2) +
geom_violin() +
ggtitle("Responder")+
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))
ggplot(deconvolution_violin_Non_Responder, aes(x=Cell_Type, y=Values, fill=Cell_Type), las = 2) +
geom_violin() +
ggtitle("Non_Responder")+
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))
ggplot(deconvolution_violin_Control, aes(x=Cell_Type, y=Values, fill=Cell_Type), las = 2) +
geom_violin() +
ggtitle("Control")+
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))
DNA methylation state
Most variable CpGs
All samples
Focus_global_methylation(match_hit_CpGs_Blueprint_EPIC, anno_EPIC, Phenotype_Wang_Feng, BMIQ_Wang_Feng_high_variance, c("Baseline", "Control"), c("#0000FF", "#555555", "#FF0000", "#00FF00"))
Baseline Cluster1 vs Cluster2
Focus_global_methylation(match_hit_CpGs_Blueprint_EPIC, anno_EPIC, Phenotype_Wang_Feng_Baseline_clustered, BMIQ_Baseline_high_variance, c("cluster1", "cluster2"), c("#0000FF", "#FF0000", "#FF0000", "#00FF00"))
Baseline Cluster1 vs Responder
cluster1_sample <- Phenotype_Wang_Feng_Baseline_clustered %>%
dplyr::filter(., Phenotype == "cluster1") %>%
.$Sample_number
BMIQ_cluster1 <- BMIQ_Baseline %>%
dplyr::select(., cluster1_sample)## Note: Using an external vector in selections is ambiguous.
## i Use `all_of(cluster1_sample)` instead of `cluster1_sample` to silence this message.
## i See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
## This message is displayed once per session.Responder_sample <- Phenotype_Wang_Feng %>%
dplyr::filter(., Phenotype == "Responder") %>%
.$Sample_number
BMIQ_Responder <- BMIQ_Wang_Feng %>%
dplyr::select(., Responder_sample)
BMIQ_cluster1_Responder <- cbind(BMIQ_cluster1, BMIQ_Responder)
Phenotype_Responder_cluster1 <- data.frame("Sample" = c(cluster1_sample, Responder_sample), "Phenotype" = c(rep("cluster1", length(cluster1_sample)), rep("Responder", length(Responder_sample))))
BMIQ_cluster1_Responder_high_variance <- Focus_high_variance_CpGs(BMIQ_cluster1_Responder)
Focus_global_methylation(match_hit_CpGs_Blueprint_EPIC, anno_EPIC, Phenotype_Responder_cluster1, BMIQ_cluster1_Responder_high_variance, c("cluster1", "Responder"), c("#0000FF", "#00FF00", "#FF0000", "#00FF00"))
Baseline Cluster2 vs Non-Responder
cluster2_sample <- Phenotype_Wang_Feng_Baseline_clustered %>%
dplyr::filter(., Phenotype == "cluster2") %>%
.$Sample_number
BMIQ_cluster2 <- BMIQ_Baseline %>%
dplyr::select(., cluster2_sample)## Note: Using an external vector in selections is ambiguous.
## i Use `all_of(cluster2_sample)` instead of `cluster2_sample` to silence this message.
## i See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
## This message is displayed once per session.Non_Responder_sample <- Phenotype_Wang_Feng %>%
dplyr::filter(., Phenotype == "Non_Responder") %>%
.$Sample_number
BMIQ_Non_Responder <- BMIQ_Wang_Feng %>%
dplyr::select(., Non_Responder_sample)
BMIQ_cluster2_Non_Responder <- cbind(BMIQ_cluster2, BMIQ_Non_Responder)
Phenotype_Non_Responder_cluster2 <- data.frame("Sample" = c(cluster2_sample, Non_Responder_sample), "Phenotype" = c(rep("cluster2", length(cluster2_sample)), rep("Non_Responder", length(Non_Responder_sample))))
BMIQ_cluster2_Non_Responder_high_variance <- Focus_high_variance_CpGs(BMIQ_cluster2_Non_Responder)
Focus_global_methylation(match_hit_CpGs_Blueprint_EPIC, anno_EPIC, Phenotype_Non_Responder_cluster2, BMIQ_cluster2_Non_Responder_high_variance, c("cluster2", "Non_Responder"), c("#0000FF", "#FF0000", "#FF0000", "#00FF00"))
Promoter of genes of interest
Non-Responder vs Responder
Focus_Gene_global_methylation("CEBPA", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng, BMIQ_Wang_Feng, c("Non-Responder", "Responder"), c("#0000FF", "#555555", "#FF0000", "#00FF00"))## Warning: Computation failed in `stat_signif()`:
## valeur manquante là où TRUE / FALSE est requis
Focus_Gene_global_methylation("CPT1A", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng, BMIQ_Wang_Feng, c("Non-Responder", "Responder"), c("#0000FF", "#555555", "#FF0000", "#00FF00"))## Warning: Computation failed in `stat_signif()`:
## valeur manquante là où TRUE / FALSE est requis
Focus_Gene_global_methylation("CPT2", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng, BMIQ_Wang_Feng, c("Non-Responder", "Responder"), c("#0000FF", "#555555", "#FF0000", "#00FF00"))## Warning: Computation failed in `stat_signif()`:
## valeur manquante là où TRUE / FALSE est requis
Focus_Gene_global_methylation("SLC25A20", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng, BMIQ_Wang_Feng, c("Non-Responder", "Responder"), c("#0000FF", "#555555", "#FF0000", "#00FF00"))## Warning: Computation failed in `stat_signif()`:
## valeur manquante là où TRUE / FALSE est requis
Focus_Gene_global_methylation("PPARGC1A", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng, BMIQ_Wang_Feng, c("Non-Responder", "Responder"), c("#0000FF", "#555555", "#FF0000", "#00FF00"))## Warning: Computation failed in `stat_signif()`:
## valeur manquante là où TRUE / FALSE est requis
Focus_Gene_global_methylation("AKT2", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng, BMIQ_Wang_Feng, c("Non-Responder", "Responder"), c("#0000FF", "#555555", "#FF0000", "#00FF00"))## Warning: Computation failed in `stat_signif()`:
## valeur manquante là où TRUE / FALSE est requis
Focus_Gene_global_methylation("PTEN", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng, BMIQ_Wang_Feng, c("Non-Responder", "Responder"), c("#0000FF", "#555555", "#FF0000", "#00FF00"))## Warning: Computation failed in `stat_signif()`:
## valeur manquante là où TRUE / FALSE est requis
Baseline Cluster1 vs Cluster2
Focus_Gene_global_methylation("CEBPA", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng_Baseline_clustered, BMIQ_Baseline, c("cluster1", "cluster2"), c("#0000FF", "#FF0000", "#FF0000", "#00FF00"))
Focus_Gene_global_methylation("CPT1A", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng_Baseline_clustered, BMIQ_Baseline, c("cluster1", "cluster2"), c("#0000FF", "#FF0000", "#FF0000", "#00FF00"))
Focus_Gene_global_methylation("CPT2", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng_Baseline_clustered, BMIQ_Baseline, c("cluster1", "cluster2"), c("#0000FF", "#FF0000", "#FF0000", "#00FF00"))
Focus_Gene_global_methylation("SLC25A20", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng_Baseline_clustered, BMIQ_Baseline, c("cluster1", "cluster2"), c("#0000FF", "#FF0000", "#FF0000", "#00FF00"))
Focus_Gene_global_methylation("PPARGC1A", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng_Baseline_clustered, BMIQ_Baseline, c("cluster1", "cluster2"), c("#0000FF", "#FF0000", "#FF0000", "#00FF00"))
Focus_Gene_global_methylation("AKT2", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng_Baseline_clustered, BMIQ_Baseline, c("cluster1", "cluster2"), c("#0000FF", "#FF0000", "#FF0000", "#00FF00"))
Focus_Gene_global_methylation("PTEN", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng_Baseline_clustered, BMIQ_Baseline, c("cluster1", "cluster2"), c("#0000FF", "#FF0000", "#FF0000", "#00FF00"))
Neighbor of promoter of genes of interest
Non-Responder vs Responder
Focus_gene_promoter_neighborhood_methylation("CEBPA", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng, BMIQ_Wang_Feng, c("Non-Responder", "Responder"), c("#0000FF", "#555555", "#FF0000", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)## Warning: Computation failed in `stat_signif()`:
## valeur manquante là où TRUE / FALSE est requis
Focus_gene_promoter_neighborhood_methylation("CPT1A", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng, BMIQ_Wang_Feng, c("Non-Responder", "Responder"), c("#0000FF", "#555555", "#FF0000", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)## Warning: Computation failed in `stat_signif()`:
## valeur manquante là où TRUE / FALSE est requis
Focus_gene_promoter_neighborhood_methylation("CPT2", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng, BMIQ_Wang_Feng, c("Non-Responder", "Responder"), c("#0000FF", "#555555", "#FF0000", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)## Warning: Computation failed in `stat_signif()`:
## valeur manquante là où TRUE / FALSE est requis
Focus_gene_promoter_neighborhood_methylation("SLC25A20", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng, BMIQ_Wang_Feng, c("Non-Responder", "Responder"), c("#0000FF", "#555555", "#FF0000", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)## [1] "No Cpgs found in neighbor of promoter of SLC25A20"## [1] FALSEFocus_gene_promoter_neighborhood_methylation("PPARGC1A", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng, BMIQ_Wang_Feng, c("Non-Responder", "Responder"), c("#0000FF", "#555555", "#FF0000", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)## [1] "No Cpgs found in neighbor of promoter of PPARGC1A"## [1] FALSEFocus_gene_promoter_neighborhood_methylation("AKT2", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng, BMIQ_Wang_Feng, c("Baseline", "Control"), c("#0000FF", "#555555", "#FF0000", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)
Focus_gene_promoter_neighborhood_methylation("PTEN", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng, BMIQ_Wang_Feng, c("Baseline", "Control"), c("#0000FF", "#555555", "#FF0000", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)
Baseline Cluster1 vs Cluster2
Focus_gene_promoter_neighborhood_methylation("CEBPA", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng_Baseline_clustered, BMIQ_Baseline, c("cluster1", "cluster2"), c("#0000FF", "#FF0000", "#FF0000", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)
Focus_gene_promoter_neighborhood_methylation("CPT1A", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng_Baseline_clustered, BMIQ_Baseline, c("cluster1", "cluster2"), c("#0000FF", "#FF0000", "#FF0000", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)
Focus_gene_promoter_neighborhood_methylation("CPT2", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng_Baseline_clustered, BMIQ_Baseline, c("cluster1", "cluster2"), c("#0000FF", "#FF0000", "#FF0000", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)
Focus_gene_promoter_neighborhood_methylation("SLC25A20", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng_Baseline_clustered, BMIQ_Baseline, c("cluster1", "cluster2"), c("#0000FF", "#FF0000", "#FF0000", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)## [1] "No Cpgs found in neighbor of promoter of SLC25A20"## [1] FALSEFocus_gene_promoter_neighborhood_methylation("PPARGC1A", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng_Baseline_clustered, BMIQ_Baseline, c("cluster1", "cluster2"), c("#0000FF", "#FF0000", "#FF0000", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)## [1] "No Cpgs found in neighbor of promoter of PPARGC1A"## [1] FALSEFocus_gene_promoter_neighborhood_methylation("AKT2", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng_Baseline_clustered, BMIQ_Baseline, c("cluster1", "cluster2"), c("#0000FF", "#FF0000", "#FF0000", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)
Focus_gene_promoter_neighborhood_methylation("PTEN", match_hit_CpGs_Blueprint_promoter_EPIC, anno_promoter_EPIC, Phenotype_Wang_Feng_Baseline_clustered, BMIQ_Baseline, c("cluster1", "cluster2"), c("#0000FF", "#FF0000", "#FF0000", "#00FF00"), matchit_CpGs_Pchic_EPIC, pchic, match_hit_CpGs_Blueprint_EPIC)
Gene Ontology
Non Responder vs Responder
BMIQ_wang_feng_analysis <- Differential_analysis(Phenotype_Wang_Feng$Phenotype, BMIQ_Wang_Feng)## 1 done## 2 done## 3 done## 4 done## 5 done## 6 doneResponse_diff_hypermet_in_Non_Responder <- BMIQ_wang_feng_analysis[["Non_Responder-Responder"]] %>%
dplyr::filter(., P.Value < 0.01) %>%
dplyr::filter(., logFC > 0.3)
Response_diff_hypermet_in_Responder <- BMIQ_wang_feng_analysis[["Non_Responder-Responder"]] %>%
dplyr::filter(., P.Value < 0.01) %>%
dplyr::filter(., logFC < 0.3)
Genes_hypermetylated_in_Non_responder <- Look_at_gene_with_CpGs_in_promoter(Response_diff_hypermet_in_Non_Responder$ID, match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_Non_responder_enhancer <- Look_at_genes_connected_to_CpGs(Response_diff_hypermet_in_Non_Responder$ID,
matchit_CpGs_Pchic_EPIC,
pchic,
matchit_Blueprint_Pchic,
match_hit_CpGs_Blueprint_promoter_EPIC)
Response_diff_hypermet_in_Baseline <- BMIQ_wang_feng_analysis[["Baseline-Control"]] %>%
dplyr::filter(., P.Value < 0.01) %>%
dplyr::filter(., logFC > 0.3)
Response_diff_hypermet_in_Control <- BMIQ_wang_feng_analysis[["Baseline-Control"]] %>%
dplyr::filter(., P.Value < 0.01) %>%
dplyr::filter(., logFC < 0.3)
Genes_hypermetylated_in_Baseline <- Look_at_gene_with_CpGs_in_promoter(Response_diff_hypermet_in_Baseline$ID, match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_Baseline_enhancer <- Look_at_genes_connected_to_CpGs(Response_diff_hypermet_in_Control$ID,
matchit_CpGs_Pchic_EPIC,
pchic,
matchit_Blueprint_Pchic,
match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_Non_responder_ego <- enrichGO(
gene = Genes_hypermetylated_in_Non_responder,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Non_responder_enhancer_ego <- enrichGO(
gene = Genes_hypermetylated_in_Non_responder_enhancer,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Baseline_ego <- enrichGO(
gene = Genes_hypermetylated_in_Baseline,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Baseline_enhancer_ego <- enrichGO(
gene = Genes_hypermetylated_in_Baseline_enhancer,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Non_responder_promoter_and_enhancer_ego <- enrichGO(
gene = unique(c(Genes_hypermetylated_in_Non_responder, Genes_hypermetylated_in_Non_responder_enhancer)),
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Baseline_promoter_and_enhancer_ego <- enrichGO(
gene = unique(c(Genes_hypermetylated_in_Baseline, Genes_hypermetylated_in_Baseline_enhancer)),
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
dotplot(Genes_hypermetylated_in_Non_responder_ego, showCategory = 30, title = "Hypermet NR")
dotplot(Genes_hypermetylated_in_Non_responder_enhancer_ego, showCategory = 30, title = "Hypermet Neighborhood NR")
dotplot(Genes_hypermetylated_in_Non_responder_promoter_and_enhancer_ego, showCategory = 30, title = "Prom & enhancer NR")
dotplot(Genes_hypermetylated_in_Baseline_ego, showCategory = 30, title = "Hypermet Baseline")
dotplot(Genes_hypermetylated_in_Baseline_enhancer_ego, showCategory = 30, title = "Hypermet Neighborhood Baseline")
dotplot(Genes_hypermetylated_in_Baseline_promoter_and_enhancer_ego, showCategory = 30, title = "Prom & enhancer Baseline")
Baseline Cluster1 vs Cluster 2
BMIQ_Baseline_analysis <- Differential_analysis(Phenotype_Wang_Feng_Baseline_clustered$Phenotype, BMIQ_Baseline)## 1 doneResponse_diff_hypermet_in_Cluster1 <- BMIQ_Baseline_analysis[["cluster1-cluster2"]] %>%
dplyr::filter(., P.Value < 0.01) %>%
dplyr::filter(., logFC > 0.3)
Response_diff_hypermet_in_Cluster2 <- BMIQ_Baseline_analysis[["cluster1-cluster2"]] %>%
dplyr::filter(., P.Value < 0.01) %>%
dplyr::filter(., logFC < 0.3)
Genes_hypermetylated_in_Cluster1 <- Look_at_gene_with_CpGs_in_promoter(Response_diff_hypermet_in_Cluster1$ID, match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_Cluster1_enhancer <- Look_at_genes_connected_to_CpGs(Response_diff_hypermet_in_Cluster1$ID,
matchit_CpGs_Pchic_EPIC,
pchic,
matchit_Blueprint_Pchic,
match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_Cluster2 <- Look_at_gene_with_CpGs_in_promoter(Response_diff_hypermet_in_Cluster2$ID, match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_Cluster2_enhancer <- Look_at_genes_connected_to_CpGs(Response_diff_hypermet_in_Cluster2$ID,
matchit_CpGs_Pchic_EPIC,
pchic,
matchit_Blueprint_Pchic,
match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_Cluster1_ego <- enrichGO(
gene = Genes_hypermetylated_in_Cluster1,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Cluster1_enhancer_ego <- enrichGO(
gene = Genes_hypermetylated_in_Cluster1_enhancer,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Cluster2_ego <- enrichGO(
gene = Genes_hypermetylated_in_Cluster2,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Cluster2_enhancer_ego <- enrichGO(
gene = Genes_hypermetylated_in_Cluster2_enhancer,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Cluster1_promoter_and_enhancer_ego <- enrichGO(
gene = unique(c(Genes_hypermetylated_in_Cluster1, Genes_hypermetylated_in_Cluster1_enhancer)),
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Cluster2_promoter_and_enhancer_ego <- enrichGO(
gene = unique(c(Genes_hypermetylated_in_Cluster2, Genes_hypermetylated_in_Cluster2_enhancer)),
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
#dotplot(Genes_hypermetylated_in_Cluster1_ego, showCategory = 30, title = "Hypermet C1")
dotplot(Genes_hypermetylated_in_Cluster1_enhancer_ego, showCategory = 30, title = "Hypermet Neighborhood C1")
dotplot(Genes_hypermetylated_in_Cluster1_promoter_and_enhancer_ego, showCategory = 30, title = "Prom & enhancer C1")
dotplot(Genes_hypermetylated_in_Cluster2_ego, showCategory = 30, title = "Hypermet C2")
dotplot(Genes_hypermetylated_in_Cluster2_enhancer_ego, showCategory = 30, title = "Hypermet Neighborhood C2", font.size = 9)
dotplot(Genes_hypermetylated_in_Cluster2_promoter_and_enhancer_ego, showCategory = 30, title = "Prom & enhancer C2")
Baseline vs Control and Non-Responder vs Responder
Genes_HM_Baseline_Non_Responder <- unique(intersect(Genes_hypermetylated_in_Non_responder, Genes_hypermetylated_in_Baseline))
Genes_HM_Baseline_Non_Responder_enhancer <- unique(intersect(Genes_hypermetylated_in_Non_responder_enhancer, Genes_hypermetylated_in_Baseline_enhancer))
Genes_HM_Baseline_Non_Responder_Prom_enhancer <- unique(c(Genes_HM_Baseline_Non_Responder, Genes_HM_Baseline_Non_Responder_enhancer))
Genes_HM_Baseline_Non_Responder_ego <- enrichGO(
gene = Genes_HM_Baseline_Non_Responder,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_HM_Baseline_Non_Responder_enhancer_ego <- enrichGO(
gene = Genes_HM_Baseline_Non_Responder_enhancer,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_HM_Baseline_Non_Responder_Prom_enhancer_ego <- enrichGO(
gene = Genes_HM_Baseline_Non_Responder_Prom_enhancer,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
dotplot(Genes_HM_Baseline_Non_Responder_ego, showCategory = 30, title = "Hypermet Base&NR")
dotplot(Genes_HM_Baseline_Non_Responder_enhancer_ego, showCategory = 30, title = "Hypermet enhancer Base&NR")
dotplot(Genes_HM_Baseline_Non_Responder_Prom_enhancer_ego, showCategory = 30, title = "Hypermet prom/enhancer Base&NR")
Baseline cluster1 vs Responder
BMIQ_Baseline_cluster1_Responder_analysis <- Differential_analysis(Phenotype_Responder_cluster1$Phenotype, BMIQ_cluster1_Responder)## 1 doneResponse_diff_hypermet_in_Cluster1_vs_Responder <- BMIQ_Baseline_cluster1_Responder_analysis[["cluster1-Responder"]] %>%
dplyr::filter(., P.Value < 0.01) %>%
dplyr::filter(., logFC > 0.3)
Response_diff_hypermet_in_Responder_vs_Cluster1 <- BMIQ_Baseline_cluster1_Responder_analysis[["cluster1-Responder"]] %>%
dplyr::filter(., P.Value < 0.01) %>%
dplyr::filter(., logFC < 0.3)
Genes_hypermetylated_in_Cluster1_vs_Responder <- Look_at_gene_with_CpGs_in_promoter(Response_diff_hypermet_in_Cluster1_vs_Responder$ID, match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_Cluster1_vs_Responder_enhancer <- Look_at_genes_connected_to_CpGs(Response_diff_hypermet_in_Cluster1_vs_Responder$ID,
matchit_CpGs_Pchic_EPIC,
pchic,
matchit_Blueprint_Pchic,
match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_Responder_vs_Cluster1 <- Look_at_gene_with_CpGs_in_promoter(Response_diff_hypermet_in_Responder_vs_Cluster1$ID, match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_Responder_vs_Cluster1_enhancer <- Look_at_genes_connected_to_CpGs(Response_diff_hypermet_in_Responder_vs_Cluster1$ID,
matchit_CpGs_Pchic_EPIC,
pchic,
matchit_Blueprint_Pchic,
match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_Cluster1_vs_Responder_ego <- enrichGO(
gene = Response_diff_hypermet_in_Cluster1_vs_Responder,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Cluster1_vs_Responder_enhancer_ego <- enrichGO(
gene = Genes_hypermetylated_in_Cluster1_vs_Responder_enhancer,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Responder_vs_Cluster1_ego <- enrichGO(
gene = Genes_hypermetylated_in_Responder_vs_Cluster1,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Responder_vs_Cluster1_enhancer_ego <- enrichGO(
gene = Genes_hypermetylated_in_Responder_vs_Cluster1_enhancer,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Cluster1_vs_Responder_promoter_and_enhancer_ego <- enrichGO(
gene = unique(c(Response_diff_hypermet_in_Cluster1_vs_Responder, Genes_hypermetylated_in_Cluster1_vs_Responder_enhancer)),
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Responder_vs_Cluster1_promoter_and_enhancer_ego <- enrichGO(
gene = unique(c(Genes_hypermetylated_in_Responder_vs_Cluster1, Genes_hypermetylated_in_Responder_vs_Cluster1_enhancer)),
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
#dotplot(Genes_hypermetylated_in_Cluster1_vs_Responder_ego, showCategory = 30, title = "Hypermet C1")
dotplot(Genes_hypermetylated_in_Cluster1_vs_Responder_enhancer_ego, showCategory = 30, title = "Hypermet Neighborhood C1", font.size = 9)
#dotplot(Genes_hypermetylated_in_Cluster1_vs_Responder_promoter_and_enhancer_ego, showCategory = 30, title = "Prom & enhancer C1", font.size = 9)
dotplot(Genes_hypermetylated_in_Responder_vs_Cluster1_ego, showCategory = 30, title = "Hypermet RES", font.size = 6)
dotplot(Genes_hypermetylated_in_Responder_vs_Cluster1_enhancer_ego, showCategory = 30, title = "Hypermet Neighborhood RES", font.size = 9)
dotplot(Genes_hypermetylated_in_Responder_vs_Cluster1_promoter_and_enhancer_ego, showCategory = 30, title = "Prom & enhancer RES")
Baseline cluster2 vs Non-Responder
BMIQ_Baseline_cluster2_Non_Responder_analysis <- Differential_analysis(Phenotype_Non_Responder_cluster2$Phenotype, BMIQ_cluster2_Non_Responder)## 1 doneResponse_diff_hypermet_in_cluster2_vs_Non_Responder <- BMIQ_Baseline_cluster2_Non_Responder_analysis[["cluster2-Non_Responder"]] %>%
dplyr::filter(., P.Value < 0.01) %>%
dplyr::filter(., logFC > 0.3)
Response_diff_hypermet_in_Non_Responder_vs_cluster2 <- BMIQ_Baseline_cluster2_Non_Responder_analysis[["cluster2-Non_Responder"]] %>%
dplyr::filter(., P.Value < 0.01) %>%
dplyr::filter(., logFC < 0.3)
Genes_hypermetylated_in_cluster2_vs_Non_Responder <- Look_at_gene_with_CpGs_in_promoter(Response_diff_hypermet_in_cluster2_vs_Non_Responder$ID, match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_cluster2_vs_Non_Responder_enhancer <- Look_at_genes_connected_to_CpGs(Response_diff_hypermet_in_cluster2_vs_Non_Responder$ID,
matchit_CpGs_Pchic_EPIC,
pchic,
matchit_Blueprint_Pchic,
match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_Non_Responder_vs_cluster2 <- Look_at_gene_with_CpGs_in_promoter(Response_diff_hypermet_in_Non_Responder_vs_cluster2$ID, match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_Non_Responder_vs_cluster2_enhancer <- Look_at_genes_connected_to_CpGs(Response_diff_hypermet_in_Non_Responder_vs_cluster2$ID,
matchit_CpGs_Pchic_EPIC,
pchic,
matchit_Blueprint_Pchic,
match_hit_CpGs_Blueprint_promoter_EPIC)
Genes_hypermetylated_in_cluster2_vs_Non_Responder_ego <- enrichGO(
gene = Response_diff_hypermet_in_cluster2_vs_Non_Responder,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_cluster2_vs_Non_Responder_enhancer_ego <- enrichGO(
gene = Genes_hypermetylated_in_cluster2_vs_Non_Responder_enhancer,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Non_Responder_vs_cluster2_ego <- enrichGO(
gene = Genes_hypermetylated_in_Non_Responder_vs_cluster2,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Non_Responder_vs_cluster2_enhancer_ego <- enrichGO(
gene = Genes_hypermetylated_in_Non_Responder_vs_cluster2_enhancer,
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_cluster2_vs_Non_Responder_promoter_and_enhancer_ego <- enrichGO(
gene = unique(c(Response_diff_hypermet_in_cluster2_vs_Non_Responder, Genes_hypermetylated_in_cluster2_vs_Non_Responder_enhancer)),
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
Genes_hypermetylated_in_Non_Responder_vs_cluster2_promoter_and_enhancer_ego <- enrichGO(
gene = unique(c(Genes_hypermetylated_in_Non_Responder_vs_cluster2, Genes_hypermetylated_in_Non_Responder_vs_cluster2_enhancer)),
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_EPIC
)
#dotplot(Genes_hypermetylated_in_cluster2_vs_Non_Responder_ego, showCategory = 30, title = "Hypermet C2")
dotplot(Genes_hypermetylated_in_cluster2_vs_Non_Responder_enhancer_ego, showCategory = 30, title = "Hypermet Neighborhood C2")
#â—‹dotplot(Genes_hypermetylated_in_cluster2_vs_Non_Responder_promoter_and_enhancer_ego, showCategory = 30, title = "Prom & enhancer C2")
dotplot(Genes_hypermetylated_in_Non_Responder_vs_cluster2_ego, showCategory = 30, title = "Hypermet NR")
dotplot(Genes_hypermetylated_in_Non_Responder_vs_cluster2_enhancer_ego, showCategory = 30, title = "Hypermet Neighborhood NR")
dotplot(Genes_hypermetylated_in_Non_Responder_vs_cluster2_promoter_and_enhancer_ego, showCategory = 30, title = "Prom & enhancer NR", font.size = 9)
Volcano plots
Non Responder vs Responder
volcanoplot_methylation(BMIQ_wang_feng_analysis[["Non_Responder-Responder"]], match_hit_CpGs_Blueprint_EPIC, "Non_Responder vs Responder")
Baseline Cluster1 vs Cluster 2
volcanoplot_methylation(BMIQ_Baseline_analysis[["cluster1-cluster2"]], match_hit_CpGs_Blueprint_EPIC, "Cluster1 vs Cluster2")
Baseline cluster1 vs Responder
volcanoplot_methylation(BMIQ_Baseline_cluster1_Responder_analysis[["cluster1-Responder"]], match_hit_CpGs_Blueprint_EPIC, "Cluster1 vs Responder")
Baseline cluster2 vs Non-Responder
volcanoplot_methylation(BMIQ_Baseline_cluster2_Non_Responder_analysis[["cluster2-Non_Responder"]], match_hit_CpGs_Blueprint_EPIC, "Cluster2 vs Non Responder")
Gene expression
Global data
Heatmaps
Baseline vs Relapse
ann_color_RNA <- list(
Phenotype = c(Baseline = "blue", Relapse = "red"))
Make_heatmap(RNAseq_Wang_Feng, Phenotype_Wang_Feng_RNAseq, "pearson", "Baseline & Relapse", ann_color_RNA)
Make_heatmap(RNAseq_Wang_Feng, Phenotype_Wang_Feng_RNAseq, "spearman", "Baseline & Relapse", ann_color_RNA)
Gene ontology
Baseline vs Relapse
gene_universe_RNAseq_Wang_Feng <- rownames(RNAseq_Wang_Feng) %>%
str_split(., "[|]") %>%
unlist(.) %>%
.[grep("[A-Za-z]", .)]
RNAseq_Wang_Feng_analysis <- Differential_analysis(Phenotype_Wang_Feng_RNAseq$Phenotype, RNAseq_Wang_Feng, "gene expression")## Warning: Zero sample variances detected, have been offset away from zero## 1 doneBaseline_Relapse_gene_expression_analysis <- RNAseq_Wang_Feng_analysis[["Baseline-Relapse"]] %>%
dplyr::filter(., )
up_entrez_in_Baseline <- Baseline_Relapse_gene_expression_analysis %>%
dplyr::filter(., logFC > 0 & P.Value < 0.05) %>%
separate_rows(., ID, sep = "[|]") %>%
dplyr::filter(., grepl("[A-Za-z]", ID))
down_entrez_in_Baseline <- Baseline_Relapse_gene_expression_analysis %>%
dplyr::filter(., logFC < 0 & P.Value < 0.05) %>%
separate_rows(., ID, sep = "[|]") %>%
dplyr::filter(., grepl("[A-Za-z]", ID))
Genes_repressed_during_Relapse <- enrichGO(
gene = unique(up_entrez_in_Baseline$ID),
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_RNAseq_Wang_Feng
)
Genes_overexpressed_during_Relapse <- enrichGO(
gene = unique(down_entrez_in_Baseline$ID),
keyType = "SYMBOL",
OrgDb = "org.Hs.eg.db",
ont = "ALL",
pAdjustMethod = "none",
universe = gene_universe_RNAseq_Wang_Feng
)
dotplot(Genes_repressed_during_Relapse, showCategory = 30, title = "Genes Down Relapse")
dotplot(Genes_overexpressed_during_Relapse, showCategory = 30, title = "Genes Up Relapse")
Volcano plot
Baseline vs Relapse
volcanoplot_gene_expression(RNAseq_Wang_Feng_analysis[["Baseline-Relapse"]], "Baseline vs Relapse")